Re: [PATCH v2] mm/swapfile: unuse_pte can map random data if swap read fails

[David Hildenbrand <david@redhat.com>]

On 16.04.22 05:05, Miaohe Lin wrote:
> There is a bug in unuse_pte(): when swap page happens to be unreadable,
> page filled with random data is mapped into user address space. In case
> of error, a special swap entry indicating swap read fails is set to the
> page table. So the swapcache page can be freed and the user won't end up
> with a permanently mounted swap because a sector is bad. And if the page
> is accessed later, the user process will be killed so that corrupted data
> is never consumed. On the other hand, if the page is never accessed, the
> user won't even notice it.
> 
> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
> ---
> v2:
>   use special swap entry to avoid permanently mounted swap
>   free the bad page in swapcache
> ---
>  include/linux/swap.h    |  7 ++++++-
>  include/linux/swapops.h | 10 ++++++++++
>  mm/memory.c             |  5 ++++-
>  mm/swapfile.c           | 11 +++++++++++
>  4 files changed, 31 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/swap.h b/include/linux/swap.h
> index d112434f85df..03c576111737 100644
> --- a/include/linux/swap.h
> +++ b/include/linux/swap.h
> @@ -55,6 +55,10 @@ static inline int current_is_kswapd(void)
>   * actions on faults.
>   */
>  
> +#define SWAP_READ_ERROR_NUM 1
> +#define SWAP_READ_ERROR     (MAX_SWAPFILES + SWP_HWPOISON_NUM + \
> +			     SWP_MIGRATION_NUM + SWP_DEVICE_NUM + \
> +			     SWP_PTE_MARKER_NUM)

Does anything speak against reusing the hwpoison marker? At least from a
program POV it's similar "the previously well defined content at this
user space address is no longer readable/writable".

I recall that we can just set the pfn to 0 for the hwpoison marker.

There is e.g., check_hwpoisoned_entry() and it just stops if it finds
"pfn=0".


-- 
Thanks,

David / dhildenb